U.S. patent number 6,455,854 [Application Number 09/529,090] was granted by the patent office on 2002-09-24 for infrared radiation detector for monitoring the presence of alkanes.
This patent grant is currently assigned to Zellweger Analytics Limited. Invention is credited to Lee Richman.
United States Patent |
6,455,854 |
Richman |
September 24, 2002 |
Infrared radiation detector for monitoring the presence of
alkanes
Abstract
The present invention relates to the infrared detection of
hydrocarbon gases; infrared light from a source (10) is passed
through a filter (12) to produce a beam (14) passing through a
space (16), which can potentially contain hydrocarbon gases. The
wavelength of the beam (14) contains a wavelength that is absorbed
by hydrocarbon gases (sample wavelength) and a wavelength that is
not absorbed by hydrocarbon gases (reference wavelength). The beam
(14) falls on a detector that contains sensors (22, 24) that
receive light that has passed through respective filters (18, 20).
Sample filter (20) allows a single wavelength to be transmitted;
reference filter (18) allows two wavelength bands to be
transmitted, the bands having wavelengths located on either side of
the sample wavelength in order to eliminate the effects of
atmospheric conditions that are unconnected with hydrocarbon
gases.
Inventors: |
Richman; Lee (Poole,
GB) |
Assignee: |
Zellweger Analytics Limited
(Poole, GB)
|
Family
ID: |
26312417 |
Appl.
No.: |
09/529,090 |
Filed: |
July 24, 2000 |
PCT
Filed: |
October 06, 1998 |
PCT No.: |
PCT/GB98/02991 |
371(c)(1),(2),(4) Date: |
July 24, 2000 |
PCT
Pub. No.: |
WO99/19712 |
PCT
Pub. Date: |
April 22, 1999 |
Foreign Application Priority Data
|
|
|
|
|
Oct 10, 1997 [GB] |
|
|
9721608 |
Oct 10, 1997 [GB] |
|
|
9721609 |
|
Current U.S.
Class: |
250/343;
250/339.01 |
Current CPC
Class: |
G01N
21/314 (20130101); G01N 21/359 (20130101); G01N
21/3504 (20130101) |
Current International
Class: |
G01N
21/31 (20060101); G01N 21/35 (20060101); G01J
005/02 () |
Field of
Search: |
;250/343,339.01,336.1,345,340,341,373 ;356/437,411,409 |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
|
|
|
|
|
|
|
744615 |
|
Nov 1995 |
|
EP |
|
1402301 |
|
Aug 1975 |
|
GB |
|
1402302 |
|
Aug 1975 |
|
GB |
|
2008745 |
|
Jun 1979 |
|
GB |
|
2163251 |
|
Feb 1986 |
|
GB |
|
Primary Examiner: Dang; Hung Xuan
Attorney, Agent or Firm: Andrus, Sceales, Starke &
Sawall, LLP
Claims
What is claimed is:
1. An infrared gas detector, said detector monitoring for the
presence of one or more of a plurality of target gases comprising
volatile, potentially explosive C.sub.1-7 alkane compounds having
differing infrared absorption spectra, said detector accurately
indicating the concentration of such gases in service environments
exhibiting differential attenuation of infrared radiation; said
detector comprising: an infrared radiation source for transmitting
at least one beam of infrared radiation having a wavelength range
that includes wavelengths of substantially 2215, 2300, and 2385 nm;
a first filter spaced from said radiation source along a beam path
in which the one or more target gases are received and which is
subjected to the service environment of the detector, said first
filter receiving radiation from the beam traversing the beam path,
said first filter transmitting radiation in a band having a central
wavelength of 2300.+-.20 nm and a full width half maximum of
50.+-.20 nm to form a first wavelength band for said detector; a
dual bandpass, second filter spaced from said radiation source
along said beam path, said second filter receiving radiation from
the beam and transmitting radiation in a band having a central
wavelength of 2215.+-.20 nm and a full width half maximum of
25.+-.20 nm to form a second wavelength band for said detector,
said second filter further transmitting radiation in a band having
a central wavelength of 2385.+-.20 nm and a full width half maximum
of 25.+-.20 nm to form a third wavelength band for said detector; a
first radiation sensor for sensing the intensity of radiation in
said first wavelength band; a second radiation sensor for sensing
the intensity of radiation in said second and third wavelength
bands; and means responsive to said first and second radiation
sensors for indicating the presence of one or more target
gases.
2. An infrared gas detector as claimed in claim 1, wherein said
first filter transmits a first wavelength band having a central
wavelength of 2300 .+-.5 nm and said dual band pass second filter
transmits second and third wavelength bands having central
wavelengths of 2215.+-.5 nm and 2385.+-.5 nm, respectively.
3. An infrared gas detector as claim in claim 1, wherein the
infrared radiation source is a Xenon arc flashlamp.
4. An infrared gas detector as claimed in claim 1, wherein said
first filter transmits a first wavelength band having a full width
half maximum of 50.+-.10 nm.
5. An infrared gas detector as claimed in claim 1, wherein said
dual band pass second filter transmits a second wavelength band
having a full width half maximum of 25.+-.5 nm.
6. An infrared gas detector as claimed in claim 1, wherein said
dual band pass second filter transmits a third wavelength band
having a full width half maximum of 25.+-.5 nm.
7. A method for monitoring for the presence of one or more of a
plurality of target gases comprising volatile, potentially
explosive C.sub.1-7 alkane compounds using infrared radiation, the
gases having differing absorption spectra, said method accurately
indicating the concentration of such gases in service environments
exhibiting differential attenuation of infrared radiation; said
method comprising the steps of: transmitting at least one beam of
infrared radiation having a wavelength range that includes
wavelengths of substantially 2215, 2300, and 2385 nm, the beam
being transmitted along a beam path in which one or more target
gases are received and which is subjected to a service environment;
forming a first wavelength band from the radiation traversing the
beam path, said first wavelength band having a central wavelength
of 2300.+-.20 nm and a full width half maximum of 50.+-.20 nm;
forming a second wavelength band from the radiation traversing the
beam path, said second wavelength band having a central wavelength
of 2215.+-.20 nm and a full width half maximum of 25.+-.20 nm;
forming a third wavelength band from the radiation traversing the
beam path, said third wavelength band having a central wavelength
of 2385.+-.20 nm and a full width half maximum of 25.+-.20 nm;
sensing the intensity of the radiation in said first wavelength
band; sensing the intensity of the radiation in said second and
third wavelength bands; and indicating the presence of one or more
target gases from the sensed intensities of the radiation in said
first, second, and third wavelength bands.
8. A method as claimed in claim 7, wherein the step of forming the
first wavelength band is further defined as forming a first
wavelength band having a central wavelength of 2300.+-.5 nm, and
wherein the steps of forming the second and third wavelength bands
are further defined as forming a second wavelength band having a
central wavelength of 2215.+-.5 nm and, a third wavelength band
having a central wavelength of 2385.+-.5 nm.
9. A method as claimed in claim 7, wherein the transmitting step is
further defined as providing the infrared radiation by a Xenon arc
flashlamp.
10. A method as claimed in claim 7, wherein the step of forming the
first wavelength band is further defined as forming a first
wavelength band having a full width maximum of 50.+-.10 nm.
11. A method as claimed in claim 7, wherein the step of forming
said second wavelength band is further defined as forming a second
wavelength band having a full width half maximum of 25.+-.5 nm.
12. A method as claimed in claim 7, wherein the step of forming
said third wavelength band is further defined as forming a third
wavelength band having a full width half maximum of 25.+-.5 nm.
Description
INDUSTRIAL FIELD
The present invention relates to the infrared detection of
hydrocarbon gases (which term includes vapours).
BACKGROUND ART
The use of non-dispersive infrared spectroscopy to detect gases is
well established. It essentially involves transmitting infrared
radiation along a path in an area being monitored; the wavelength
of the infrared radiation is chosen so that it is absorbed by the
gas of interest (hereafter called the "target gas") but not
substantially absorbed by other gases in the atmosphere of the area
being monitored. If monitoring out-of-doors, the wavelength should
ideally not be absorbed by liquid water (e.g. in the form of
condensation, rain or spray). The intensity of the radiation that
has passed along the path in the area being monitored is measured
and the attenuation in the intensity of the radiation gives a
measure of the amount of the target gas in the monitored area.
However, factors other than absorption by the target gas also
attenuate the infrared radiation including obscuration of the
detecting beam, atmospheric scattering of the radiation,
contamination of optical surfaces, e.g. by dirt or condensation,
and ageing of components. The reliability of infrared gas detectors
is significantly improved by the use of a reference; such a
reference is usually infrared radiation at a different wavelength,
which ideally is a wavelength at which the target gas does not
exhibit significant absorption. The ratio between the signal
obtained at the wavelength where the target gas does absorb (the
"sample wavelength") and the signal obtained at the wavelength
where the target gas does not significantly absorb (the "reference
wavelength") compensates for the attenuation caused by non-target
gases since ideally the signal at the reference wavelength and the
signal at the sample wavelength will both be affected by such
non-target gas attenuation.
A known infrared detector is a so-called "fixed point" detector,
which has a very short path length (e.g. up to 10 cm) and so only
monitors a relatively small space. It can be used to detect
leakages of hydrocarbons from oilrigs, pipelines, storage tanks or
refineries. The provision of such detectors in open spaces away
from a leakage site may result in the leakage not being detected
since prevailing atmospheric conditions (e.g. wind speed, wind
direction and temperature) could carry the gas away from the
detectors, which would then not register the leakage. It is
therefore a difficult task to position such fixed point detectors
and usually a compromise is drawn in the location of detectors,
based on likely leak sites and typical prevailing weather
conditions; also the number of such detectors that can be provided
is limited by cost. Generally fixed-point detectors are used for
monitoring of specific items of equipment and apparatus that are
liable to leak, for example pipeline joints and valves.
Fixed-point detectors are coupled with an alarm that indicates the
detection of a target gas in the immediate neighbourhood of the
detector. Because such detectors are placed near the source of any
leak, any significant leaking target gas will be in relatively high
concentration in and around the detector. It is therefore possible
to set the alarm such that the amount of the target gas present
before the alarm is triggered is relatively large, thereby avoiding
the giving of false alarms. The giving of false alarms is a
substantial problem since it could result in the shutting down of a
facility, for example an oil rig or an oil refinery.
To overcome the above-mentioned shortcomings of fixed-point
detectors, longer path-length gas detectors, so called "open-path
optical gas detectors", are used, in which radiation at sample and
reference wavelengths is transmitted along an open-path which
passes through the atmosphere in the space to be monitored. The
length of the path can vary from one to a thousand meters,
depending on the application, and so allows a much greater space to
be monitored than is the case with fixed-point detectors. When used
out-of-doors, the open nature of the optical path means that the
beam is exposed to prevailing atmospheric weather conditions, which
can seriously affect the operation of the instrument. For example,
rain, snow, mist, fog, sea spray, blizzards and sand or dust storms
scatter or absorb radiation at the reference and sample
wavelengths. The level of absorption and scattering by such weather
conditions depends on the size, shape, nature and optical
properties of the droplets, drops or particles concerned.
Unfortunately such attenuation is not uniform across the infrared
spectrum, i.e. the attenuation at the sample wavelength and the
attenuation at the reference wavelength are not identical which
gives rise to errors in the measurement of the amount of target gas
and can, in extreme cases, lead to the failure to trigger an alarm
or the triggering of a false alarm. The matter is complicated
considerably because different weather conditions exhibit different
relative and absolute attenuation at the sample and reference
wavelengths. For example, one sort of fog can attenuate the
radiation at the reference wavelength more than the radiation at
the sample wavelength whereas a different sort of fog will
attenuate the radiation at the sample wavelength more than the
radiation at the reference wavelength. The variability of
atmospheric attenuation for different weather conditions makes it
very difficult to compensate for the effects of weather upon this
sort of gas detector.
In order to minimise the differential attenuation between the
sample wavelength and the reference wavelength, it is preferable
that the two wavelengths are as close as possible to each other.
However, this is not always possible since there may not be a
suitable reference wavelength, i.e. a wavelength at which the
target gas is only minimally absorbed, near the sample wavelength.
The situation is made even more complex because of the need to
avoid cross-sensitivity to other atmospheric gases, the
absorption/refraction characteristics of water droplets that may be
present in the path of the infrared beam and the band shapes and
tolerances of the filters used to restrict the sample and reference
wavelengths. Thus there may be several hundred nanometers between
the sample and reference wavelengths. This separation can result in
significant differences between the absorption characteristics of
the sample and reference wavelengths under different weather
conditions, as set out above.
Instead of measuring the reference signal at a single wavelength,
it has been proposed (see for example GB-1,402,301, GB-1,402,302,
U.S. Pat. No. 4,567,366, EP-0,744,615 and GB-2,163,251) to use two
reference wavelengths located on either side of a sample wavelength
and take, as the reference signal, the average of the signals at
the two reference wavelengths. This arrangement requires the
measurement of light absorption at two different reference
wavelengths, which in turn requires either the use of separate
light beams to measure the absorption at each of the two reference
wavelengths or the use of a mechanical arrangement to bring two
filters into alignment with a single light-sensitive detector. Both
solutions will work satisfactorily in a laboratory but not in the
field, particularly not in the harsh environments encountered on
offshore oil/gas platforms or in the Middle East, the Tropics, the
Arctic, etc. The use of two reference light beams (in addition to
the sample light beam) requires careful alignment (within micron
tolerances) of the detectors and it is difficult enough to align
the detectors for the sample beam and a single reference beam, let
along aligning an additional detector for a second reference beam.
Furthermore, the buffeting of the detectors in the environment of
the North Sea, for example, can displace the alignment. In
addition, the use of an additional reference beam makes the system
expensive. The use of a mechanical arrangement (e.g. a spinning
filter wheel) to bring sample and reference filters periodically
into alignment with a single light-sensitive detector is also not
feasible in the field since vibrations can affect the operation of
the mechanical arrangement and such mechanical parts can be
unreliable in the harsh environmental conditions encountered.
It is also known to use as the reference a broad range of
wavelengths, and if the range includes the sample wavelength, it is
possible to make the average reference wavelength equal to the
sample wavelength, thereby eliminating the problems discussed above
of having the reference and sample wavelengths distant from each
other. However, the inclusion in the reference signal of a
substantial component made up of the sample signal itself leads to
substantially reduced sensitivity.
All the C.sub.1-7 alkanes are gaseous or volatile and their escape
can give rise to a risk of an explosion and therefore it is
necessary to monitor for the presence of any of them. It is not
feasible to provide separate systems for detecting each alkane and
therefore it is necessary that a single system should be capable of
detecting all these alkanes. However, the alkanes all have
different spectra and so it is difficult to select a single sample
wavelength and a single reference wavelength that can be used for
all of the C.sub.1-7 alkanes.
A further substantial problem underlying the use of this type of
detector is the need to avoid false alarms being given, which can
result in the shut down of a complete facility, for example oil
pipeline, oil refinery or oil rig. It is obviously desirable to be
able to detect the smallest possible concentration of the target
gas that could give rise to a hazard. However, this must be set
against the need to avoid false alarms and the above-described
problem of atmospheric conditions differentially affecting the
reference and sample wavelengths.
It is an object of the present invention to provide an infrared
detector of the open path type described that is capable of
monitoring for all, of the C.sub.1-7 alkanes and yet is
sufficiently rugged that it can be operated reliably in the field
in harsh environments. It is a further object of the present
invention to provide an infrared detector that will have an
improved accuracy for detecting alkanes in a variety of harsh
weather conditions as compared to known open path infrared
detectors so that the instances of false alarms are reduced.
DISCLOSURE OF THE INVENTION
According to the present invention there is provided an infrared
gas detector for a target gas comprising: an infrared source
capable of transmitting at least one infrared beam comprising
radiation in at least one wavelength band at which the target gas
is absorbent (sample wavelength band) and in at least one
wavelength band at which the target gas is only absorbent to a
substantially lesser extent than at the sample wavelength band
(reference wavelength band); a first radiation intensity sensor
capable of sensing the intensity of the infrared radiation in the
sample wavelength band(s) and a second radiation intensity sensor
capable of sensing the intensity of the infrared radiation in the
reference wavelength band(s), both of which are spaced apart from
the transmitter by a beam path; and a first filter that is located
in front of the first sensor and that only transmits one or more
sample wavelength bands and a second filter that is located in
front of the second sensor and that only transmits one or more
reference wavelength bands, wherein the aggregate number of sample
and reference wavelength bands transmitted by the two filters is at
least three and wherein the centre of the sample wavelength band
(if there is only one sample wavelength band) or the mid point
between the sample wavelength bands (if there are multiple sample
wavelength bands) is approximately the same as the centre of the
reference wavelength band (if there is only one reference
wavelength band) or the mid point between reference wavelength
bands (if there are multiple reference wavelength bands).
By making the mid-point of the reference wavelenlgth band(s)
approximately the same as the mid-point of the sample wavelength
band(s), the differential attenuation of radiation at the sample
and reference wavelengths caused by atmospheric conditions can be
eliminated or substantially reduced. As indicated, the aggregate
number of sample and reference wavelength bands transmitted by the
two filters can exceed three so long as the mid point between the
bands at the sample wavelengths is approximately the same as the
mid point of the bands at the reference wavelengths, although exact
coincidence of the two mid points is not required.
A single infrared beam emitted by the transmitter can contain the
sample and the reference wavelength bands or separate beams may be
provided. Indeed, the beam emitted by the transmitter can include a
wide spectrum of wavelengths, not only the sample and reference
wavelength bands. A filter may be inserted in the infrared beam in
order to transmit along the detection path only wavelengths around
the wavelengths detected by the detector sensor(s) but such a
filter is not necessary. The use of two separate radiation
intensity sensors, one for the sample wavelength band(s) and the
other for the reference wavelength band(s) means that there are
only two sensors in the system that need aligning and there is no
need to move the first and second filters.
In order to discriminate between infrared radiation from the
transmitter and infrared radiation from other potentially
interfering sources of radiation, e.g. sunlight, the radiation from
the transmitter is preferably modulated with a distinct
characteristic that can be recognised by the receiver; this
modulation can be achieved in many ways, including pulse or
amplitude modulation of the infrared source's drive
voltage/current, acousto-optic modulation and electro-optic
modulation. In the preferred embodiment, the modulation is achieved
by pulsing the voltage applied to a flashlamp, e.g. a Xenon arc
flashlamp. This pulsing produces short, very high intensity pulses
that are easily discriminated from both natural and artificial
sources that are likely to be encountered in the intended operating
environment.
One highly important aspect of the present invention is the
selection of the sample and reference wavelength bands that allow
the reliable detection of C.sub.1-7 alkanes. The first (sample)
filter advantageously transmits a single band having a central
wavelength of 2300.+-.5 nm and a full width half maximum (FWHM) of
50.+-.10 nm. The second (reference) filter is preferably a dual
bandpass filter having a first band centred around 2215.+-.5 nm
with a FWHM of 25.+-.5 nm and a second band having a central
wavelength of 2385.+-.5 nm and a FWHM of 25.+-.5 nm. It can be seen
that the sample wavelength band lies mid way between the two
reference wavelength bands of the dual bandpass reference filter.
The selection of these wavelengths is not apparent from the spectra
of C.sub.1-7 alkanes since some of the alkanes are significantly
absorbent at the reference wavelengths and yet, as shown below, we
have surprisingly shown that the choice of these wavelengths
provides good detection of C.sub.1-7 alkanes and avoids giving
false alarms even in the presence of rain and fog.
DETAILED DESCRIPTION OF DRAWINGS
The present invention will be further described by way of example
only with reference to the following drawings in which:
FIG. 1 is a schematic graph showing the change in the transmission
characteristics of the beam path without any target gas being
present as a function of wavelength in the presence of two
different types of atmospheric conditions 1 and 2;
FIG. 2 is a schematic graph of the intensity of infrared radiation
in the sample and reference bands in the prior art arrangement
without any target gas being present, as modified by environmental
condition 1 of FIG. 1;
FIG. 3 is a schematic graph of the intensity of infrared radiation
in the sample and reference bands in the prior art arrangement
without any target gas being present, as modified by environmental
condition 2 of FIG. 1;
FIG. 4 is a schematic graph of the intensity of infrared radiation
in the sample and reference bands according to the present
invention without any target gas being present, as modified by
environmental condition 1 of FIG. 1;
FIG. 5 is a schematic graph of the intensity of infrared radiation
in the sample and reference bands according to the present
invention without any target gas being present, as modified by
environmental condition 2 of FIG. 1; and
FIG. 6 is a schematic view of the detector of the type used in the
present invention and in the prior art. The difference between the
present invention and the prior art lies in the nature of the
filters, as explained in further detail below.
DETAILED DISCLOSURE OF PREFERRED EMBODIMENTS
Referring initially to FIG. 6, there is shown a source of infrared
radiation, which in the present invention is preferably a flash
lamp 10, placed behind a filter 12 that transmits infrared
radiation in sample and reference wavelength bands, as described in
further detail below. The beam 14 transmitted by filter 12 passes
along beam path 16, which can vary in length from one to a thousand
meters, and is attenuated by material in that path. The beam 14
attenuated in this way is incident on a receiver 17 containing a
beam splitter 19; filters 18 and 20 and corresponding intensity
measuring sensors 22 and 24 are placed in the two parts of the
split beams. Filter 18 transmits radiation only at the sample
wavelengths and filter 20 transmits radiation only at the reference
wavelengths.
In prior art arrangements, filter 20 transmits in a single sample
wavelength band and the filter 20 transmits in a single reference
wavelength band as described below with reference to FIGS. 2 and 3.
In the exemplified detector of the present invention filter 18 is a
single bandpass filter transmitting radiation in one distinct
source infrared band and filter 20 is a dual bandpass filter
transmitting radiation in two distinct reference infrared bands
that lie either side of the sample band of filter 18, as described
below with reference to FIGS. 4 and 5.
The signals from sensors 22 and 24 are fed into a circuit 26 that
can calculate the ratio between these two signals. If there is
target gas in the path of the beam 14, radiation in the sample
wavelength band is absorbed by the target gas whereas radiation in
the reference wavelength band is not (or at least not to the same
extent). The ratio of the intensity of the signal of sensor 22
(measuring the sample wavelengths) as compared to the signal of
sensor 24 (measuring the reference wavelengths) will therefore fall
in the presence of a target gas and if the ratio falls below a
threshold value, an alarm 28 is triggered indicating the presence
of the target gas. The threshold value can be set by the user.
Referring to FIG. 1, which shows the transmittance (T) of the
signal from the radiation source 10 and filter 12, as attenuated by
environmental conditions 1 and 2 prevailing along beam path 16. As
can be seen, under environmental condition 1, the transmittance of
the beam is greater at higher wavelengths than at lower wavelengths
whereas the situation is completely the other way round under
environmental conditions 2; as mentioned above, fog can constitute
both environmental condition 1 and environmental condition 2,
depending on the size of the water droplets in the fog. Referring
to FIG. 2, there is shown a plot of radiation intensity against
wavelength; in the known prior art open path infrared gas
detectors, a single sample wavelength band A and a single reference
wavelength band B are used. The intensity of radiation in the
sample wavelength band A incident on sensor 22 and the intensity of
radiation in the reference wavelength band B incident on sensor 24
are shown. The upper curves, a' and b', show the intensities that
are unattenuated by environmental conditions 1 or 2. However, the
lower curves, a" and b" show the resulting curves when
environmental condition 1 is prevailing. As can be seen from FIG.
2, under environmental conditions 1, the reference intensity (band
B) is greater than the sample intensity (band A) even though there
is no target gas in the beam path; such a relationship between the
two signals is the usual indicator that there is some target gas in
the space being monitored. Thus, environmental conditions 1 can
lead to a premature triggering of the alarm 28 indicating the
presence of a hydrocarbon when, in fact, that is not the case.
Referring now to FIG. 3, it can be seen that the attenuation of the
sample and reference signals under environmental conditions 2 mean
that the sample intensity a" is greater than the reference
intensity b". Therefore, the presence of a greater amount of
hydrocarbon gas is required in order to trigger off the alarm
signal under environmental conditions 2 than would normally be the
case.
Referring to FIG. 4, the corresponding conditions are shown using
the detector according to the present invention. In this
embodiment, a single sample wavelength band D is used, which is
situated between (and preferably mid-way between) two reference
wavelength bands C and E. Under environmental condition 1. the
intensity of the radiation c" in reference band C is smaller than
the intensity d" at sample wavelength band D which in turn is
smaller than the intensity e" at reference wavelength band E.
However, since the signal produced by intensity sensor 24 results
from the combined intensities c" and e" of reference bands C and E,
the "average" attenuation on the two reference bands C and E under
environmental condition 1 is approximately the same as the
attenuation of wavelength band D, i.e. the differential attenuation
at different wavelengths is cancelled out and so the ratio of the
signal at sensor 22 resulting from sample wavelength band D to the
signal at sensor 24 resulting from the combined intensities at
wavelength bands C and E properly reflects the amount of target gas
in the beam path.
Referring to FIG. 5, there is shown the corresponding position
resulting from environmental condition 2. Again, the ratio of the
signal at sensor 22 resulting from sample wavelength band D to the
signal at sensor 24 resulting from the combined intensities at
wavelength bands C and E properly reflects the amount of target gas
in the beam path and is more or less the same as that prevailing in
environmental condition 1.
The above system is predicated on the assumption that the
attenuation brought about by different environmental conditions is
linear across the infrared spectrum. Whereas this is rarely, if
ever, the case, we have found that the attenuation brought about by
environmental conditions can generally be approximated to linear.
Whether the attenuation is exactly linear or not, it can be seen
that, by comparing the bands in FIGS. 4 and 5 with those of FIGS. 2
and 3, the present invention provides a significant benefit over
the prior art.
The selection of the sample and the reference wavelength bands will
depend on the nature of the target gas and the nature of other
gases in the space being monitored and can be selected by an
analysis of the infrared spectra of the gases involved, preferably
also taking into account the spectrum of liquid water. They should
be chosen such that the mid-point of the sample wavelengths and the
mid-point of the reference wavelengths are as close to each other
as possible.
EXAMPLE
We have found that the reference filter 20 for the detection of
C.sub.1-7 alkane hydrocarbon gases, while avoiding the effects of
water, is preferably a dual bandpass filter having a first band
centred around 2215.+-.nm (preferably 2215.+-.10 nm and more
preferably 2215.+-.5 nm) with full width half maximum (FWHM) of
25.+-.20 nm (preferably 25.+-.10 nm and more preferably 25.+-.5
nm). The second band of filter 20 has a central wavelength of
2385.+-.20 nm (preferably 2385.+-.10 nm and more preferably
2385.+-.5 nm) and a FWHM of 25.+-.20 nm (preferably 25.+-.10 nm and
more preferably 25.+-.5 nm).
The sample filter 18 transmits a single band having a central
wavelength of 2300.+-.20 nm (preferably 2300.+-.10 nm and more
preferably 2300.+-.5 nm) and a FWHM of 50.+-.20 nm (preferably
50.+-.10 nm). Thus it can be seen that the sample wavelength band
lies mid way between the reference wavelength bands of dual
bandpass filter 20.
In order to demonstrate the effect of the present invention, two
open path gas detectors were used. One was a conventional
Searchline Excel open path gas detector obtainable from Zellweger
Analytics (Hatch Pond House, 4 Stinsford Road, Nuffield Estate,
Poole, Dorset BH1 0RZ, United Kingdom) fitted with conventional,
single bandpass sample and reference filters, the other was an
identical Searchline Excel detector but modified by using different
sample and reference filters so that the detector was in accordance
with the present invention. The filters used were narrow bandpass
sample filter and a double bandpass reference filter. The centre
wavelengths and the full width half maximum values of the filters
of the two detectors are set out in Table 1.
TABLE 1 Centre Wavelength(s) FWHM Conventional Detector Sample
2,300 nm .+-. 25 nm 200 nm .+-. 25 nm Reference 2,100 nm .+-. 25 nm
200 nm .+-. 25 nm Detector according to the Present Invention
Sample (Narrowband): 2,300 nm .+-. 5 nm 50 nm .+-. 5 nm Reference
(Double bandpass) 2,215 nm .+-. 5 nm 25 nm .+-. 5 nm & &
2,385 nm .+-. 5 nm 25 nm .+-. 5 nm
The two gas detectors were tested by aligning each detector such
that their beams passed through a tunnel in which controlled,
simulated fog conditions could be produced. The detectors were then
zeroed and calibrated using a standard procedure.
Whilst monitoring the gas reading (measured in lower explosive
limit (LEL).m) from the detectors under test, the density of fog in
the tunnel was gradually increased from zero up to an attenuation
of 2,000:1. At regular intervals, the detectors were functionally
tested using plastic test filters, which simulate a nominal 3.0
LEL.m gas cloud.
It was found that the detector fitted with the conventional filters
showed a steady increase in gas reading as the fog density
increased. This reading reached 0.5 LEL.m at an attenuation of
40:1. The detector continued to function up to 2,000:1 attenuation
but with a significantly offset zero.
The detector unit fitted with the filters of the present invention
showed no significant gas reading as the fog density was increased.
The detector continued to function acceptably up to 2.000:1
attenuation. The only change in behaviour noticed at high
attenuations was a slight increase in response to the plastic
functional test filter.
The detector unit fitted with conventional filters exhibited a
significant response to fog. The first alarm threshold for most
applications in which open path gas detectors are used is generally
set at 1.0 LEL.m. In order to ensure that false alarms are not
generated by environmental conditions, a maximum zero offset of 0.5
LEL.m is allowable. The detector fitted with conventional filters
exceeded this for fog attenuations of 40:1 (approximately 25 meters
visibility). This imposes a restriction on the density of fog in
which the detector can be allowed to remain operational.
The detector fitted with the new design of filters exhibited no
significant response to fog up to attenuations of 2,000:1. The
slight increase in test filter response was not significant and was
in the safe direction (i.e. increased the output). The performance
enhancement achieved with the use of the new filters enables high
integrity operation in fog conditions with attenuation up to at
least 500:1. This is sufficient to enable operation in all but the
most extreme fog conditions, which are rare enough to represent no
significant operational problem.
* * * * *